{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:JE6GW6NG22O25IT4W5GMRHH5WS","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"f17ecd3e826c661d245a8e9bc85c8b3997ab160eda0eab0196cbd94b3098edad","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.AI","submitted_at":"2026-06-03T16:36:54Z","title_canon_sha256":"6acded4d5985f77417cefa082c8bc20039bd1498f4c87d6770ab4517a2dcd0ba"},"schema_version":"1.0","source":{"id":"2606.05080","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2606.05080","created_at":"2026-06-04T01:10:05Z"},{"alias_kind":"arxiv_version","alias_value":"2606.05080v1","created_at":"2026-06-04T01:10:05Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.05080","created_at":"2026-06-04T01:10:05Z"},{"alias_kind":"pith_short_12","alias_value":"JE6GW6NG22O2","created_at":"2026-06-04T01:10:05Z"},{"alias_kind":"pith_short_16","alias_value":"JE6GW6NG22O25IT4","created_at":"2026-06-04T01:10:05Z"},{"alias_kind":"pith_short_8","alias_value":"JE6GW6NG","created_at":"2026-06-04T01:10:05Z"}],"graph_snapshots":[{"event_id":"sha256:336d75d48052a6db0e86ed915f7d41a81c98986ac8e530d065c4b782645db815","target":"graph","created_at":"2026-06-04T01:10:05Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2606.05080/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Scientific and engineering progress is fundamentally a long-horizon iterative process: proposing changes, running experiments, measuring outcomes, and continuously refining artifacts. Yet existing benchmarks for frontier models primarily evaluate either single-turn responses or short-horizon agent trajectories, failing to capture the challenges of sustained iterative improvement over extended time horizons. To address this gap, we introduce AutoLab, a new benchmark for ultra long-horizon closed-loop optimization. AutoLab consists of 36 realistic, expert-curated tasks spanning four diverse doma","authors_text":"Alex Pentland, Dongfu Jiang, Hang Hua, Hao Zhang, Jiaxin Pei, Jiefeng Chen, Jinsung Yoon, Junda Chen, Lichi Li, Mengdi Wang, Misha Sra, Radha Poovendran, Shizhe Diao, Yue Huang, Zexue He, Zhangchen Xu, Zheyuan Liu, Zichen Chen, Zijian Wu","cross_cats":["cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.AI","submitted_at":"2026-06-03T16:36:54Z","title":"AutoLab: Can Frontier Models Solve Long-Horizon Auto Research and Engineering Tasks?"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.05080","kind":"arxiv","version":1},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:451f1961d8b4c97ac83710a1d5a39613db2c4723e7d1c4d31934aaa67a8c10ff","target":"record","created_at":"2026-06-04T01:10:05Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"f17ecd3e826c661d245a8e9bc85c8b3997ab160eda0eab0196cbd94b3098edad","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.AI","submitted_at":"2026-06-03T16:36:54Z","title_canon_sha256":"6acded4d5985f77417cefa082c8bc20039bd1498f4c87d6770ab4517a2dcd0ba"},"schema_version":"1.0","source":{"id":"2606.05080","kind":"arxiv","version":1}},"canonical_sha256":"493c6b79a6d69daea27cb74cc89cfdb48482fa94d958d1c5ce58f57f196bc0d6","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"493c6b79a6d69daea27cb74cc89cfdb48482fa94d958d1c5ce58f57f196bc0d6","first_computed_at":"2026-06-04T01:10:05.485642Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-04T01:10:05.485642Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"KO93vX8lopXXYBYHjg1pV+pyQ9Dh/8OsLsdIPuen5yF5lk06azgtm51byqx3aFhomRUPxGs4cGimWKzVclJaAg==","signature_status":"signed_v1","signed_at":"2026-06-04T01:10:05.486411Z","signed_message":"canonical_sha256_bytes"},"source_id":"2606.05080","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:451f1961d8b4c97ac83710a1d5a39613db2c4723e7d1c4d31934aaa67a8c10ff","sha256:336d75d48052a6db0e86ed915f7d41a81c98986ac8e530d065c4b782645db815"],"state_sha256":"bdebc6c145e67839dafa2b1d7ecb343d5d7a9285e57b8c16377933b56760f5a8"}